Field of the Invention
Embodiments of the present invention generally relate to angular intra-prediction in video coding.
Description of the Related Art
The Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T WP3/16 and ISO/IEC JTC 1/SC 29/WG 11 is currently developing the next-generation video coding standard referred to as High Efficiency Video Coding (HEVC). Similar to previous video coding standards such as H.264/AVC, HEVC is based on a hybrid coding scheme using block-based prediction and transform coding. First, the input signal is split into rectangular blocks that are predicted from the previously decoded data by either motion compensated (inter) prediction or intra prediction. The resulting prediction error is coded by applying block transforms based on an integer approximation of the discrete cosine transform, which is followed by quantization and entropy coding of the transform coefficients.
The above compression process is inherently lossy. While the quantization of the transform coefficients compresses the video by taking advantage of perceptual redundancy in the video, it inevitably introduces quantization errors. In some real world applications, such lossy coding is undesirable. For example, in automotive vision applications, video captured from cameras in a vehicle may need to be transmitted to central processors in a lossless manner for purposes of applying video analytics. In another example, in web collaboration and remote desktop sharing applications where hybrid natural and syntactic video coding might be required, part of the video scene may contain synthetic contents such as presentation slides as well as graphical representation of function keys in a user interface that need to be losslessly coded.